OBSEEVATIONS ON SOME SOUTH AFRICAN TERMITES. 459 



with belli cosus, or whether under the circumstances one 

 can select a portion of the Haviland series as a variety^ I am 

 not in a position to say; but it can be said that the differences 

 cited for the two species in the Monogi-aph (2) are not of 

 specific value. 



Some particulars regarding my examinations of T . natal- 

 ensis may not be out of place here. 



P'orty series examined composed as follows : 



Fourteen nest collections represented by soldiers, workers, 

 and imag-os (kings and queens). From Natal, three (Mount 

 Edgecombe, Bellair, Winterton). From the Transvaal, 

 eleven (Pretoria, Pietersburg, Barbei-ton). 



Seven nest collections represented by soldiers and workers 

 only. From Natal, four (Pietermaritzbiirg, New Hanover, 

 Weeueu). From the Transvaal, three (Pretoria, Tzaneen). 



Twelve nest collections represented by two or more soldiers 

 from each (Haviland material from the district of Plstcourt, 

 Natal). 



Seven field collections represented by soldiers and workers. 

 From Natal, two (Bellair, Pietermaritzburg). From the Trans- 

 vaal, five (Pretoria and district). 



An examination of the series of major soldiers before me 

 brings out one or two peculiar features. The first is that, as 

 regards colour, markings, and general appearance, the longer 

 material has been in alcohol the more do the specimens 

 harmonise. Thus whilst Haviland's insects, which have been 

 in spirit since 1898, are all agreeably alike ; colour notes, 

 made from certain material, when freshly preserved twelve 

 luonths ago, do not wholly apply at the present moment. 

 Again, many colour markings and features are more readily 

 seen upon submerged specimens than dry ones. On the other 

 hand certain other features may be obscured, more particularly 

 the sculpturing of the frons. The appearance of the insects 

 is changed in alcohol as regards shape as well as colour ; the 

 whole insect becomes clarified, and the inflation of the abdo- 

 men, so marked in living material, disappears, probably by 

 the slow dissolving of certain body-contents and the escape 

 of gases and air. 



VOL. 3, PART 2. 31 



